Dissecting and targeting mechanisms of genomic instability-triggered immune evasion in RBM10-deficient non-small cell lung cancer

Project Summary/Abstract Oncogenic mutations in KRAS represent the most prevalent genomic driver event in lung adenocarcinoma (LUAD) (~30%) and account for ~25,000 deaths annually in the U.S. Immunotherapy (IO) with immune checkpoint inhibitors (ICI) is active in KRAS-mutant non-small-cell lung cancer (NSCLC), however only a minority of patients exhibit long-lasting responses. Co-occurring genomic alterations can shape the immunobiology of KRAS-mutant NSCLC and impact its response to ICI. Loss-of-function somatic mutations in RBM10, encoding a regulator of alternative splicing (AS), are prevalent in LUAD (~8%) and are significantly enriched in KRAS- mutant NSCLC (~25%). We found that loss of RBM10 in KRAS-mutant NSCLC tumors and cell lines results in DNA double-strand breaks (DSBs) and STING-dependent but cGAS-independent NF-κB signaling, that establish an immunosuppressive tumor microenvironment (TME) – rich in MDSCs and M2-macrophages – and support tumor immune escape. Critically, RD NSCLC exhibit selective sensitivity to anti-CSF1R, that depends on functional CD8+ T-cells. Preliminary evidence points to accumulation of R-loops and distinct STING isoforms as candidate mechanisms that underpin DDR activation and preferential NF-κB engagement in RD cells. Based on our preliminary findings we hypothesize that: 1. In KRAS-mutant NSCLC, RBM10 loss triggers R-loop accumulation and aberrant DDR signaling that support STING-dependent but cGAS-independent pro- tumorigenic NF-κB signaling; 2. Splicing dysregulation upon RBM10 loss promotes STING isoforms that preferentially engage NF-κB over TBK1 and IRF3; 3. RBM10 loss remodels the NSCLC TME and fosters immune evasion 4. The sensitivity of RD NSCLC to ICI can be enhanced by co-targeting STAT3 with TTI-101. In Aim 1, we will dissect the link between RBM10 loss, DDR activation and STING-mediated NFκB signaling and we will assess the contribution of altered R-loop homeostasis and alternative STING splicing to these phenotypes. In Aim 2, we will comprehensively characterize the composition, signaling pathways and functional properties of the RD TME in preclinical models in order to identify critical mediators of immune evasion and we will validate key findings in NSCLC clinical specimens. Finally, in Aim 3, we will determine the impact of RBM10 inactivation on the clinical efficacy of ICI using clinical outcome data/specimens from patients enrolled in two phase 3 clinical trials of durvalumab with or without tremelimumab versus platinum-doublet chemotherapy for previously untreated metastatic NSCLC as well a phase 3 clinical trial of nivolumab/ipilimumab. In addition, we will evaluate co-targeting STAT3 in combination with anti-PD-(L)1 in order to enhance the efficacy of immune checkpoint blockade in RD-NSCLC. Clinical significance: This work will examine a novel link between splicing dysregulation and immune evasion that is mediated by STING and will further seek to develop precision combination immunotherapy approaches for RD NSCLC. The strength of our assembled multi-disciplinary team of experts will facilitate rapid translation of discoveries into clinical advances for NSCLC patients.